EP2199824B1 - Method of selecting frequencies and transmitted beam directions for a radar using a frequency dispersive antenna - Google Patents

Description

The present invention relates to the general field of multifunctional radar management systems with electronic scanning. It relates to methods for managing resources in radar time, that is to say the methods for optimally managing the sequence of the scores made. By pointing is meant the exploitation, for the implementation of the same function, of the signals obtained by transmitting a given waveform in a direction of the space given by the diagram of the antenna. The invention applies in particular to the management of clocking by a dispersive slot antenna monitoring radar.

The phenomenon of dispersive aperture is a phenomenon that affects some radars (slot antenna radars). It is reflected in a known manner by a deterministic misalignment of the antenna beam as a function of frequency. These radars are generally standby radars, mono function, the cost of implementation is relatively low.
For such radars, this phenomenon, which affects the pointing direction of the beam, is very generally considered as a disadvantage that needs to be corrected in real time. To do this, for example, the azimuth deviation caused is conventionally taken into account in the management of the scores so that, in view of this deviation, the desired waveform is transmitted in the desired direction.

In the case of a conventional radar standby and in particular in the case of a dispersive antenna standby radar, the simplicity of the operation implemented through a fixed and deterministic definition of the sequence of pointing and frequencies to during the scanning of space by the antenna. This sequence is pre-established in advance for different frequency plans (authorized and / or non-scrambled frequencies). It is generally stored in a table which is read periodically with a constant periodicity which corresponds to the theoretical rotation speed of the antenna. In this way, for a given dotted direction, the antenna being considered to have a determined and constant rotational speed, the waveform played is known deterministically from one turn to another.

In this type of radar, the good management of the scores is therefore generally dependent on the accuracy and stability of the rotational speed of the antenna. Thus, if the actual speed of rotation of the antenna is not equal to the theoretical speed, the position of the antenna at a given instant is different from that required. Moreover, the existence of a fluctuation of the rotational speed on the antenna revolution results in an uncontrolled fluctuation of the value of the difference in azimuth between adjacent beams, fluctuation which can not be regulated by the loaded means. management of radar weather resources.

With regard to dispersive antennas, however, it is known to exploit the variation of the direction of pointing of the antenna lobe caused by a variation of the transmission frequency, so as to provide the radar in question with de-positioning capabilities. making it possible to perform radar transmissions and plays in directions having a given angle with respect to the axis of the antenna. The United States patent specification issued to COM DEV Ltd, numbered 4,868,574, notably describes a dispersive antenna radar comprising transmitting and receiving means enabling it to exploit this feature to realize a pointing the antenna beam in varying directions depending on the frequency used.

However, when it is desired to have a radar capable of sequencing stints, no longer in a predetermined manner but in a dynamic manner, it is known to use a radar-type multi-function radar If electronic scanning, rotating antenna, equipped means for managing the scores made in the antenna beam for each azimuth and for each antenna tower. Such management is described in particular in the European patent application filed by the applicant and published under the reference EP 0645839 . However, it is particularly adapted to the operation of such radars and is not directly usable to manage the operation of a watch radar, to dispersive antenna, simpler design and not having, in particular, actual misalignment capabilities of the antenna beam.

An object of the invention is to take advantage of the dispersive phenomenon to extend the operational capabilities of a dispersive antenna standby radar by making it capable of operating to a certain extent in the manner of a multi-function radar while ensuring the diversity of frequencies played (the equiprobability and non-predictability of the frequencies used according to the authorized frequency plans), the regulation of the load (relative to the current radar time budget and the speed variation of the antenna) and the robustness of the jamming radar (robustness against instantaneous listening of scrambled frequencies and information maintained in turn on scrambled frequency charts). Another object of the invention, relating to the management of the radar time budget, is to maintain the rate of operational use of the radar, by regulating, and no longer undergoing, variations in the instantaneous spacing between the contiguous scores. when the antenna period is not at its nominal value (rotation of the antenna on average slower or faster) and that the antenna rotation speed fluctuates quite strongly in the lathe.

• une étape de sélection des pointages requis dont la direction est visible à l'instant considéré;
• une étape de sélection par les fréquences autorisées;
• une étape de sélection des fréquences éligibles les moins brouillées;
• une étape de sélection des fréquences éligibles les moins utilisées;
• une étape de création de nouvelles requêtes de pointages;

les étapes du procédé formant un cycle répété séquentiellement, la table qui associe les pointages candidats et les fréquences candidates étant actualisée à chaque cycle.For this purpose, the subject of the invention is a method for managing the emission of scores by a radar comprising a dispersive antenna whose speed is liable to vary over time, the management being carried out as a function of the angle of rotation of the antenna. the antenna, the method being applied to candidate scores, each candidate score being associated in a table with one or more candidate frequencies, characterized in that it comprises the following steps:

• a step of selecting the required scores whose direction is visible at the moment considered;
• a selection step by the authorized frequencies;
• a step of selecting the least scrambled eligible frequencies;
• a step of selecting the least used eligible frequencies;
• a step of creating new score requests;

the steps of the process forming a sequentially repeated cycle, the table which associates the candidate scores and the candidate frequencies being updated at each cycle.

In a preferred embodiment, the method according to the invention also comprises a final complementary step to deal with the case of the scores which are no longer visible at the end of the current iteration taking into account the direction of the antenna.

In a particular embodiment, the method according to the invention also comprises a first intermediate step which consists in excluding from the selection the candidate scores which, if they were finally selected during the iteration considered, could induce the loss of one or more other candidate scores whose visibility is low. This first intermediate step is placed after the first main step.

In a particular embodiment, which can be combined with the preceding, the method according to the invention further comprises a second intermediate step for selecting the scores declared as being the highest priority. This second intermediate step is placed after the second main step.

In a preferred embodiment of the method according to the invention, the step of creating new pointing requests proceeds for each candidate pointing to the association of a range of frequencies comprised in a range bounded by two frequencies f _min. and f _max . The frequency f _max is the the highest frequency frequency domain actually allocated to the radar. The frequency f _min is a randomly selected frequency in a frequency range extending from the lowest frequency of the authorized frequency plan, to a f- _{limit-stretching} frequency greater than F _min and less than f _max .

In this preferred embodiment, the frequency f _min can be obtained by a random draw which follows a decreasing density probability law when the frequency increases.

In this preferred embodiment, the frequency _{f_drawing limit} can be determined so as to define with f _max a frequency interval containing a number n of low allowed frequencies in front of the number of authorized frequencies.

In a preferred embodiment of the method according to the invention, the step of creating new pointing requests takes into account, for the determination of the new candidate scores, a first angular window contiguous to the field of visibility for the determination of new watch scores and a second angular window contiguous to the first window for the determination of the other new scores. The angular windows are determined so as to take into account the antenna rotation period and the average duration of the clockings.

In this preferred embodiment, the first angular window is determined so as to correspond to the angle of rotation of the antenna over a period equivalent to the maximum duration of a half-score, to which the delay is added. maximum that can exist between the moment when a score was selected and the moment when it was actually issued. The size of the second angular window is in turn defined as being proportional to the DAz _{AdaptationCharge} azimuth extension value which corresponds to a multiple of the azimuthal extension corresponding to the partial visibility range, DAz _{AdaptationCharge} being in all cases lower than full visibility area.

In a preferred embodiment of the method according to the invention, the step of selecting the least scrambled eligible frequencies takes into account the information relating to the least scrambled frequencies for the selection of the scores in two possible ways, either locally by instantaneous listening of scrambled frequencies either globally through the use of available scrambled frequency cards.

Advantageously, the method according to the invention makes it possible to schedule on a dispersive antenna broadband scores referred to as "Recognition of non-cooperative targets" by the optimized sequencing of a series of successive narrow-band scores Each narrow-band score is played successively in time towards the target, using the antenna rotation to play the emissions in narrow bands one after the other.

Advantageously also, the method according to the invention allows the insertion of addressed punches, tracking points for example, in the general sequencing of the scores, and this in a manner similar to a two-plane electronic scanning multifunction radar.

• les figures 1 et 2, des illustrations du problème posé dans la gestion des pointages par la présence de fluctuations temporelles non maîtrisées de la vitesse de rotation d'antenne ;
• la figure 3, une illustration des paramètres décrivant le phénomène de dispersivité;
• la figure 4, une illustration du principe d'exploitation du phénomène de dispersivité par le procédé selon l'invention;
• la figure 5, un organigramme de principe des différentes étapes du procédé de planification des pointages selon l'invention;
• la figure 6, des illustrations de différents exemples de répartition des plans de fréquences de fonctionnement d'un radar ainsi que des paramètres de gestion des fréquences associées à un pointage;
• la figure 7, une illustration de la manière dont la borne inférieure de la fenêtre des fréquences de fonctionnement dont l'utilisation est autorisée, attribuée à chaque pointage, est tirée aléatoirement;
• la figure 8, une illustration du principe de sélection des pointages à partir de leur domaine de visibilité géré et la position courante de l'antenne en azimut;
• la figure 9, une illustration de la méthode pour limiter le nombre des pointages susceptibles d'être non exécutés du fait de l'expiration des intervalles de temps durant lesquels ils sont exécutables;
• la figure 10, l'illustration de la façon dont la troisième étape du procédé selon l'invention prend en compte le caractère discret des fréquences de fonctionnement et effectue la sélection des fréquences discrètes candidates compatibles de la fréquence théorique correspondant à la direction du pointage considéré;
• la figure 11, les illustrations des différents principes de sélection d'une fréquence, pouvant être mis en oeuvre par la sixième étape du procédé selon l'invention, lorsque plusieurs fréquences sont encore candidates à ce stade;
• la figure 12, l'illustration du principe de réactualisation de la table des pointages candidats mis en oeuvre au cours de la septième étape du procédé selon l'invention.
• la figure 13, l'illustration du principe de la gestion des temps morts (instant pour lequel aucun pointage n'est jouable ou ne peut être joué) par l'insertion de pointages techniques de durée déterminée;
• la figure 14, une illustration des effets provoqués par une vitesse de rotation d'antenne différente de la valeur nominale attendue et par des fluctuations de cette vitesse de rotation.

The features and advantages of the invention will be better appreciated thanks to the description which follows, description which sets forth the invention through a particular embodiment taken as a non-limiting example and which is based on the appended figures, figures which represent:

• the Figures 1 and 2 , illustrations of the problem posed in the management of the scores by the presence of uncontrolled time fluctuations of the antenna rotation speed;
• the figure 3 , an illustration of the parameters describing the phenomenon of dispersivity;
• the figure 4 an illustration of the principle of exploitation of the phenomenon of dispersivity by the process according to the invention;
• the figure 5 a flowchart of principle of the different stages of the method of planning the clockings according to the invention;
• the figure 6 , illustrations of different examples of the distribution of the operating frequency plans of a radar as well as the parameters management of frequencies associated with a score;
• the figure 7 , an illustration of how the lower limit of the operating frequency window authorized for use, assigned to each score, is drawn randomly;
• the figure 8 , an illustration of the principle of selecting the scores from their field of view and the current position of the antenna in azimuth;
• the figure 9 , an illustration of the method for limiting the number of punishments that may be missed because of the expiration of the time intervals during which they are executable;
• the figure 10 illustrating how the third step of the method according to the invention takes into account the discrete character of the operating frequencies and selects the compatible candidate discrete frequencies of the theoretical frequency corresponding to the direction of the pointing considered;
• the figure 11 the illustrations of the different principles of selection of a frequency, which can be implemented by the sixth step of the method according to the invention, when several frequencies are still candidates at this stage;
• the figure 12 , the illustration of the principle of reactualization of the table of candidate clockings implemented during the seventh step of the method according to the invention.
• the figure 13 , the illustration of the principle of the management of idle time (instant for which no score is playable or can not be played) by the insertion of fixed-term technical scores;
• the figure 14 , an illustration of the effects caused by an antenna rotation speed different from the expected nominal value and by fluctuations in this speed of rotation.

où Δω représente un biais constant par rapport à la vitesse de rotation théorique ω₀ et où δω(t) représente un terme de fluctuation autour de la vitesse biaisée qui varie en fonction du temps c'est-à-dire en fonction de la direction vers laquelle le lobe d'antenne est orienté.
Par suite de cette variation non maitrisée de la vitesse de rotation, on peut assister à une répartition irrégulière des pointages réalisés, non maîtrisée par la gestion des faisceaux, matérialisés par les flèches 21 sur la figure 2, bien que les pointages soient, comme dans le cas de la figure 1, régulièrement commandés au cours du temps. La conséquence d'une telle variation de la vitesse de rotation de l'antenne est notamment que l'écart angulaire entre pointages n'est pas régulé. Ainsi, du fait d'un séquencement rigide initialement défini pour une vitesse de rotation constante, des fréquences non autorisées dans une direction donnée peuvent par exemple être utilisées. Cette répartition irrégulière des pointages réalisés peut occasionner des surcharges locales du séquencement des pointages radar à exécuter, surcharges que les radars classiques, du type à antenne dispersive, ne peuvent pas absorber.
La variation de la vitesse de rotation d'antenne, qui se traduit par un biais sur cette vitesse et par une fluctuation autour de ce biais, a pour conséquence que, comme l'illustre la figure 14 les pointages 141 sont plus ou moins écartées angulairement.
L'illustration 14-a correspond à une vitesse nominale avec un écart entre pointages qui correspond juste à la valeur d'atténuation à 3 dB des faisceaux correspondants (les courbes figurant la largeur du faisceau à -3 dB sont alors tangentes au point de jonction).
L'illustration 14-b correspond quant à elle, à une vitesse plus faible pour laquelle les faisceaux de pointages consécutifs 142 et 143 sont systématiquement resserrés et se superposent dans les zones correspondant aux largeurs de lobes à 3 dB.
L'illustration 14-c correspond, quant à elle, à une vitesse plus élevée pour laquelle les faisceaux de pointages consécutifs 144 et 145 sont systématiquement disjoints.
Enfin, l'illustration 14-d correspond à une situation de fluctuation de la vitesse autour d'une vitesse nominale, situation pour laquelle l'écart entre les faisceaux de pointages consécutifs, 146 et 147 ou 148 et 149, évolue de pointage à pointage.The Figures 1 and 2 schematically illustrate the effect of variations in the antenna rotation speed on the quality of coverage achieved by a standby radar whose management system of the scores does not take into account these variations.
The figure 1 , corresponds to the theoretical case where the antenna of the radar rotates at a constant speed ω ₀ precisely known. In this circumstance, he It is possible to deterministically control the moment t when the lobe of the antenna is directed in a given direction θ. As a result, it is possible to implement a standby mode of operation in which the space is regularly explored, the directions pointed, materialized by the arrows 11 on the figure 1 , being regularly spaced and precisely determined.
The figure 2 as for it, presents the case of an antenna, for which the speed of rotation is not perfectly regulated. In such a case under certain atmospheric conditions (gusty winds) the nominal rotational speed ω ₁ of the antenna is not strictly equal to the theoretical rotation speed ω ₀ . In addition, the rotational speed ω ₁ is not constant over time so that the rotation speed of the actual antenna can be written: $${\mathrm{<figure-callout\; id="1"\; label="\omega "\; filenames="imgf0006.png,imgf0008.png"\; state="\{\{state\}\}">\omega </figure-callout>}}_{\mathrm{1}}\mathrm{=}{\mathrm{<figure-callout\; id="0"\; label="\omega "\; filenames="imgf0006.png,imgf0008.png"\; state="\{\{state\}\}">\omega </figure-callout>}}_{\mathrm{0}}\mathrm{+}\mathrm{\Delta \omega }\mathrm{+}\mathrm{\delta \omega }\left(\mathrm{t}\right)$$

where Δω represents a constant bias with respect to the theoretical rotation speed ω ₀ and where δω (t) represents a fluctuation term around the biased velocity which varies as a function of time, that is to say according to the direction towards which the antenna lobe is oriented.
As a result of this uncontrolled variation in the speed of rotation, there may be an irregular distribution of the scores achieved, not controlled by the management of the beams, shown by the arrows 21 on the figure 2 , although the scores are, as in the case of the figure 1 , regularly ordered over time. The consequence of such a variation in the rotational speed of the antenna is that the angular difference between scores is not regulated. Thus, because of a rigid sequencing initially defined for a constant speed of rotation, unauthorized frequencies in a given direction can for example be used. This irregular distribution of the scores achieved may cause local overloads sequencing radar pointing to perform overloads that conventional radars, type dispersive antenna can not absorb.
The variation of the antenna rotation speed, which is reflected in a bias on this speed and in a fluctuation around this bias, has the following effect: consequence, as illustrated by figure 14 the scores 141 are more or less angularly apart.
Figure 14-a corresponds to a nominal speed with a difference between scores which corresponds to the 3 dB attenuation value of the corresponding beams (curves representing the beam width at -3 dB are then tangent to the junction point ).
Figure 14-b, for its part, corresponds to a lower speed for which the consecutive pointing beams 142 and 143 are systematically tightened and superimposed in the zones corresponding to the 3 dB lobe widths.
The illustration 14-c corresponds, meanwhile, to a higher speed for which the consecutive pointing beams 144 and 145 are systematically disjoint.
Finally, Figure 14-d corresponds to a situation of fluctuation of the speed around a nominal speed, situation for which the difference between the consecutive pointing beams, 146 and 147 or 148 and 149, evolves from point to point .

Consequently, if it is desired to control or at least regulate the positions of the directions actually pointed instantaneously, taking into account the authorized frequencies, it is necessary to set up means making it possible to compensate as much as possible for this fluctuation phenomenon. the rotational speed of the antenna. The process according to the invention advantageously constitutes such a means.

As stated above, the operating principle of the method according to the invention is based on the use of the phenomenon of azimuth dispersivity caused to the emission (and reception) diagram of an antenna, of slot antenna type by for example, by the frequency variation of the radar equipped with such an antenna. The figure 3 illustrates this phenomenon which results in the fact that, as a function of the emission frequency, the radiation pattern of such an antenna, represented by the dotted arrows 32 in the figure, has an angular offset 33 with respect to the direction antenna 31, offset whose value is a function of the transmission frequency. By antenna direction is meant here the axis perpendicular to the plane of the antenna.
This angular offset is included in a sector 34 defined by the operating frequency range of the radar. The sector 34 depends on the frequency band Δf used by the radar and depends on the minimum and maximum frequencies that can be used in the band Δf. In the example of the figure 3 , the sector 34 is shown as being shifted in advance with respect to the axis of the antenna 31.However, the sector 34 may be offset late with respect to the axis of the antenna 31. It may still be to be positioned on either side of the axis of the antenna 31 and in this case, for a slot antenna, physical constraints make the sector 34 is defined in two subsectors, right and left, separated by an unauthorized intermediate sector.

• La nécessité de tenir compte des fréquences d'émission autorisées et leur répartition dans la bande de fréquences allouée au fonctionnement du radar (cet ensemble de fréquences autorisées est constitué de fréquences discrètes qui peuvent être contiguës ou disjointes. Dans le cas de fréquences contiguës l'ensemble peut être défini par la totalité du plan des M fréquences disponibles ou par une sous-bande de N fréquences. Dans le cas de fréquences disjointes les fréquences autorisées sont définies par un peigne de P fréquences non-adjacentes, qui peut se limiter à une seule fréquence).
• la nécessité de tenir compte du fait que chaque fréquence autorisée doit être employée de façon équiprobable (en moyenne sur le tour et de tour à tour) et non prédictible (le choix d'une fréquence à un instant t ne peut être prédit par la connaissance d'un horizon fini des fréquences jouées) de façon à contrer du mieux possible certains types de brouillages.
• la nécessité de tenir compte du fait que la durée de la forme d'onde émise est variable d'un pointage à l'autre.
• la nécessité de tenir compte d'une hiérarchie dans l'importance des pointages demandés, matérialisée par un niveau de priorité accordé à chaque pointage.
• la nécessité de tenir compte de variations possibles de la vitesse de rotation de l'antenne qui conditionnent l'intervalle de temps durant lequel ce pointage peut être mis en oeuvre.
• la nécessité d'assurer qu'un nombre réduit de pointages ne sera pas exécuté;
• la nécessité de tenir compte des fréquences brouillées de façon instantanée (grâce à un temps d'écoute avant l'émission de chaque pointage) en faisant réaliser les dernières étapes de sélection du pointage par l'antenne (envoi de fréquences/pointages candidats et sélection de la fréquence selon la "Fréquence La Moins Brouillée"(FMB)).
• La nécessité de tenir compte des fréquences brouillées de tour à tour en fonction des informations de la carte des fréquences brouillées entretenue et mise à jour de tour à tour et par secteur.
• le souhait de minimiser les "temps morts" (temps de non-émission de pointages) par l'utilisation dans ces cas de pointages techniques dont le nombre dépend du budget temps radar courant.
  Dans la suite de ce document le principe des différentes fonctions mises en oeuvre pour prendre en compte ces différentes contraintes dans la planification des pointages est décrit de manière plus détaillée.

Conversely, taking as a reference angular a pointed direction, we see, as illustrated by the figure 4 , that for a given direction θ 41 and a given transmission frequency band Δf, it is possible to define an angular sector 43, of width Δθ = [θ-θ _min θ-θ _max ], such that, taking into account the rotation ω of the antenna it is always possible by playing on the transmission frequency to deflect the beam of the antenna in the direction θ while the antenna direction 42 scans this sector. As a result, it can be seen that for a given direction θ 41, a pointing can be performed in this direction, taking advantage of the deflection due to the presence of dispersivity, as soon as and as long as the direction of the antenna 42 is present during the antenna rotation in the angular interval Δθ = [θ-θ _min θ-θ _max ].
The method according to the invention puts into practice this principle to make a single-function radar capable of functioning, to a certain extent as a multi-function radar. The limitations of this ability include the maximum and minimum deflections that can be achieved using dispersivity and the minimum and maximum operating frequencies that can be used. The method according to the invention also puts into practice this principle to compensate at least in part the variations of the antenna rotation speed.
For this purpose, its main function is to temporally sequence the order in which different waveforms can be implemented, during the rotation of the antenna, each waveform to be applied for a given pointing direction and in a given period of time, knowing that a significant number of scheduling of the scores is possible because the choice at a given time of a given operating frequency makes it possible to execute a pointing rather than another. To do this, the scheduling is done taking into account, among other things, the maximum duration during which the beam of the antenna can be pointed in a given direction taking into account the instantaneous rotation speed of the antenna and the antenna. value of the angular sector Δθ, knowing that other points are to be achieved in this same time interval.
In general, the waveform to be implemented in a direction of the space covered by the radar is determined by the function (standby function or other addressed functions, such as a tracking function) that must be set implement the radar in this direction. We speak in known manner of pointing, a score corresponding to the use of a given waveform to illuminate in a manner specific to the pointing function a given direction. It should be noted that, with regard to the aiming function addressed (for example tracking tracking), this is generally managed by the global radar management unit and is reflected in the method according to the invention. by taking into account pointing requests which define the characteristics of the tracking points to be made during the rotation of the antenna (direction, associated adapted duration waveform and degree of priority of the pointing considered). On the other hand, as far as the standby function is concerned, it is directly managed by the method according to the invention, insofar as the waveform used is generally determined and the pointed directions are determined. the duration of the clockings, the estimated instantaneous rotation speed of the antenna and the radar load considered locally. By radar load is meant here the number of points to be executed in a given time interval.
As a result, the main function of the method according to the invention consists in determining, at given instant t, among the set of required scores the one to be performed at a given instant t ', situated in the near future, taking into account the position predicted antenna at this time t 'to come. The moment t 'to come considered is usually the one that corresponds to the end date of execution of the running score at the considered instant t. To carry out this task, the method according to the invention comprises different processing modules that cooperate to take into account in real time the following constraints:

• The need to take into account the permissible transmission frequencies and their distribution in the frequency band allocated to the operation of the radar (this set of authorized frequencies consists of discrete frequencies which may be contiguous or disjoint.) In the case of contiguous frequencies, can be defined by the totality of the available M frequencies plane or by a sub-band of N frequencies.In the case of disjoint frequencies the allowed frequencies are defined by a comb of P non-adjacent frequencies, which may be limited to single frequency).
• the need to take into account the fact that each authorized frequency must be used equiprobably (on average on the turn and in turn) and unpredictable (the choice of a frequency at a time t can not be predicted by the knowledge a finite horizon of frequencies played) in order to counter as well as possible certain types of interference.
• the need to take into account that the duration of the emitted waveform varies from one score to another.
• the need to take into account a hierarchy in the importance of the requested scores, materialized by a priority level granted to each score.
• the need to take into account possible variations in the speed of rotation of the antenna which condition the time interval during which this pointing can be implemented.
• the need to ensure that a reduced number of scores will not be executed;
• the need to take into account the scrambled frequencies in an instantaneous manner (thanks to a listening time before the emission of each score) by having the last steps of selection of the pointing by the antenna (sending of frequencies / candidate scores and selection the frequency according to the "Lowest Frequency" (FMB)).
• The need to take into account scrambled frequencies in turn according to the information of the scrambled frequency card maintained and update in turn and by sector.
• the desire to minimize "dead time" (time of non-emission of scores) by the use in these cases of technical scores whose number depends on the current time radar budget.
  In the remainder of this document the principle of the various functions implemented to take into account these different constraints in the planning of the scores is described in more detail.

The method according to the invention, called "management of radar pointing" (GPR) finds its place in the control chain of the radar operating modes and is positioned between the module responsible for the "management of radar tasks" (GTR) and the module responsible for "Radar Pulse Burst Management" (GRR).
The radar task management module (GTR) provides the method according to the invention in the form of task execution requests, information relating to the characteristics of certain scores or families of scores which it requires execution, tracking pointers. mainly as well as some particular standby scores (watch defined by sector). These characteristics are notably the direction of the pointing, the nature of the waveform used, the period of renewal of the scores (for the periodic scores like "the pursuit") as well as the degree of priority associated with the execution of the score considered.
The method according to the invention GPR takes into account these requests and incorporates them in a timely manner in a table that it maintains periodically over time. This table, or table of candidate scores, contains all the information relating to the scores that can be implemented during a given time interval, referred to herein as "candidate scores". The GPR method manages the scheduling of the scores recorded in the table taking into account the position of the antenna axis, the instantaneous antenna rotation speed and the usable frequencies (ie the allowed and uncuffed frequencies) by each score then delivers to the module responsible for the management of radar pulse bursts (GRR) the corresponding list of bursts to be transmitted.
As illustrated by figure 5 , the method according to the invention (GPR) is a iterative process consisting of eight successive steps 51 to 58. the two steps 52 and 54 have an optional character while the steps 51, 53, 55 to 58 constitute the essential steps of the method. Step 58 is a final step that can be associated with the process and that is intended to deal with the case of scores that have not finally been selected will no longer be visible at the next iteration, given the nominal direction of the antenna.
The method according to the invention draws up and maintains the table 50 of the candidate scores, each candidate score being initially associated, at each iteration, with one or more transmission frequencies (a transmission frequency that can be associated with several candidate scores for this step), then regularly analyzes the scores contained in the table so as to determine the order in which these candidate scores are to be implemented. In other words, its purpose is to select at each iteration one of the candidate scores contained in the table 50 and one of the candidate frequencies initially associated with this score. This pair (pointing, frequency) is the first in date to be implemented by the radar.
It should be noted that there is no bijection in the table between a score and a frequency. A candidate score may have several associated frequencies and conversely a candidate frequency may be associated with several candidate scores.
According to the invention the set of transmission frequencies associated with each score is determined from a random draw realized in the frequency plan globally attributed to the radar for its operation. In the preferred embodiment of the invention this random draw is a particular draw, whose main purpose is to ensure the diversity of frequencies that will ultimately played by the radar, all points combined. In this way, each candidate score is associated, for a given iteration, with a set of frequencies of its own, two scores can however, by chance, be associated with the same set of frequencies which represents a subset the frequency plan assigned to the radar. This subset is thus defined by a frequency f _min and a frequency f _max .

To perform the selection of a pair (pointing, frequency), each of the steps 51 to 56 applies a specific treatment to the candidate scores stored in the table and the associated frequencies, the treatment applied to each step to remove from the final selection (end of iteration) is one or more scores (steps 51, 52, 54 of the method) or one or more frequencies of the frequency plan allocated to the radar (steps 53, 55 and 56).
The processing implemented at a given stage is applied either to the selected candidate scores or to the candidate frequencies retained at the end of the preceding steps.
The scores and frequencies that are not discarded at the end of a given stage are subject to the selection of the next step, while the scores that are discarded are set aside for the remainder of the iteration so as to be optionally analyzed for again during the next iteration.

As a result the torque (pointing, frequency) finally retained at the end of the last step of the method is transmitted to the module to the management of radar pulse bursts (GRR).

In the remainder of the document, the set of eight steps that can comprise the method according to the invention is described keeping in mind that since steps 52 and 54 are optional, the method according to the invention may comprise in a simplified version only six steps.

The first step 51 consists, as illustrated by figures 6 and 8 , to determine among the candidate scores contained in the table, the scores 81, 82 whose managed range of visibility, which represents a fraction of the allocated range of visibility, does not contain the direction of the antenna at the instant considered. These scores are in principle excluded from the rest of the selection.
According to the invention, the table of candidate scores includes for each score at a time the waveform characteristics associated with the score, the level of priority of the score considered with respect to the other scores of the table, the required direction θ ₀ of the score, the duration of the score and the fraction of the visibility range allocated to the score. according to the invention defines the allocated range of visibility associated with a pointing as the azimuthal aperture accessible by deflection of the radar beam at the instant considered, given the operating frequencies available (usable) to achieve a score in the direction considered . Similarly, the managed visibility area associated with a score is also defined as the azimuthal aperture accessible by deflection of the radar beam at the instant considered, taking into account the operating frequencies (candidate frequencies) associated by taking the score considered in the candidate score table, these frequencies being selected in the range of usable frequencies. In other words, the managed visibility domain is a fraction of the allocated visibility domain. It is determined from the frequency range allocated to the score considered. This frequency range itself represents a subset of the authorized frequency plan allocated to the radar.
The allowed frequency plan can take many forms, as illustrated in Figures 6-a through 6-c of the figure 6 . It may for example consist of a set of M consecutive frequencies 61 covering the entire frequency plan B allocated to the operation of the radar (6-a) or a sub-band of N contiguous frequencies 62 (6-b) of the band of B. It may also consist (6-c) in a set of P frequencies 63 distributed disjointly or not on the whole of the band B (called gap gap comb). It will be noted here that the frequency band B allocated to the radar is very generally a set of discrete frequencies regularly spaced between a frequency F _min and a frequency F _max .
According to the invention, the frequency range assigned to each pointing request, the frequencies which form the managed visibility domain, is determined, as illustrated by the diagram of FIG. figure 6 by a minimum frequency f _min chosen by means of a random process among the frequencies constituting the authorized frequency plan B which are lower than a given frequency f _{draw limit} , and a maximum frequency f _max greater than f _{draw limit} and lower or equal to the maximum frequency F _max of this frequency plan. The set of n frequencies belonging to the interval [f _s f _max ] forms an area of the managed visibility domain called security window. According to the invention n is defined as a small number in front of the number of authorized frequencies. The principle of determination of f _min is explained in the rest of the description relating to step 57.
As a result, step 51 selects the scores considered as visible with respect to the criteria mentioned above, the other candidate scores being then discarded.
It should also be noted that, if certain candidate frequencies are associated only with candidate scores that are discarded, these frequencies are, consequently, immediately discarded from the choices made subsequently during the iteration considered. Thus, the selection made in step 51, a step which concerns in principle only the candidate scores, can in practice influence the selection of the frequencies.

According to the invention, the second step 52 is applied to the candidate scores that were not rejected at the end of the first step 51. It consists, as illustrated by FIG. figure 9 , to discard the scores 82 which, if finally selected during the iteration considered, could lead to the loss of one or more other candidate scores 81 whose deadline is close, that is to say points 81 which, if they were discarded, could no longer be selected at the next iteration, given the time required to issue the score 82 retained at the iteration considered. This is particularly the case if, as illustrated by figure 9 , the direction of the antenna is at the moment of the selection near the exit 93 of the managed visibility area of the (or) pointing (s) aside (s). The principle of this second step is therefore to take into account the execution times 91 and 92 of the scores in question 81 and 82 to determine whether the selection of a candidate score 82 operated at the current iteration may eliminate another candidate score 81 at the next iteration, this score, whose deadline is close, may or may not be a candidate score at the time of selection.
As a result of the second step 52, the selected candidate scores are either the scores already selected during step 51, or the scores considered to be priorities according to their due dates.
In order to deal with the case where several points close to their due date are likely to be lost, the process begins iteratively the tests with the highest priority of these. In case the remaining score at this stage is no playable relative to its frequency, previously deselected scores are reactivated.

The third step 53 of the method according to the invention aims to select from the scores selected at the end of the previous step, step 51 or 52 as the case may be, only the scores of which at least one of the associated frequencies corresponds to the azimuthal deflection to operate (relative to the direction of the antenna) to perform this pointing in the required direction, these frequencies being the authorized frequencies corresponding to the managed range of visibility specific to each score. The principle of this selection is illustrated by illustrations 10-a and 10-b of the figure 10 which illustrate two possible cases, the first case 10-a corresponding to a score associated with two frequencies.
To do this, the frequency or frequencies allocated to it, which correspond to the moment of selection to the required direction 101 or 102 (theoretical frequency) 101 of the score, are determined for each score, taking into account a tolerance represented by the window. 103 on illustrations 10-a and 10-b of the figure 10 . Figure 10-a presents the case where for a given score one can only retain one of the associated frequencies 105. Figure 10-b presents the case where for a given score two of the associated frequencies can be retained. 104 and 105.
The tolerance materialized by the window 103 makes it possible to limit the difference between the required direction 101 or 102 and the actually pointed direction 104 or 105 because the available frequencies are distributed in a discrete manner. The tolerance window is defined according to the performance constraints of the radar.
Consequently, if among the candidate frequencies associated with the candidate score considered, there exists at least one frequency in the vicinity of the frequency f ₀ corresponding to the theoretical direction θ ₀ of the score, that is to say an allocated frequency lying in the tolerance window then the score is retained and the frequency or frequencies close to f ₀ that allowed him to be selected
Thus, in the event that none of the frequencies allocated to the candidate score is in this window, it is discarded.
Similarly, in the hypothesis, corresponding to Figure 10-a, where a only frequency 103 among the allocated frequencies is in this window, the pointing and this frequency are retained.
Finally, assuming, corresponding to the illustration 10-b, where several of the frequencies 104 and 105 allocated are located in this window, the score and these two frequencies are retained.
It is recalled here that, as illustrated by figure 10 and as has been said previously, there is no bijective relationship between a candidate score and a candidate frequency (a candidate frequency 105 may be associated with several candidate scores, and conversely a candidate score may have several frequencies 104, 105 candidates ).

According to the invention, the fourth step 54 of the method is applied to candidate scores that were not rejected after the third step 53. It consists in taking into account the level of priority assigned to each of the candidate scores. This level of priority depends, in particular, on the nature of the score considered (for example, different priorities may be assigned according to the importance of the type of score, such as between standby and tracking scores) and the current position of the direction of travel. the antenna in its field of visibility managed (a pointing whose antenna direction enters its window backup sees its priority increase). At the end of this step, if a score is of a priority level higher than the priority levels of the other scores selected after the third step 53, this score is retained. Similarly, if several scores have an identical priority level, higher than the priority level of the other candidate scores selected after the third step 53, these scores are retained.

• Si aucune des fréquences candidates associées aux pointages candidats retenus n'est brouillée aucun des pointages candidats n'est écarté.
• si toutes les fréquences candidates sont brouillées, le pointage conservé est celui associé à la fréquence la moins brouillée.
• si seulement certaines fréquences candidates sont brouillées, seuls les pointages candidats qui ne sont associés qu'à des fréquences brouillées sont écartés.

The fifth step 55 of the method according to the invention, aims to remove from the final selection those candidate scores retained in the previous step, step 53 or 54 depending on the case, for which all the associated frequencies are declared scrambled or, in the case where the frequencies associated with the candidate scores are all scrambled, to exclude all the candidate scores other than those associated with the least scrambled frequency.
The determination of the scrambled frequencies can be carried out different known ways. It is for example possible to use a map of scrambled frequencies, also established by the radar during listening phases for example. This map is generally established for the entire radar authorized frequency plan. This map is maintained in turn by sector according to the results of the frequencies listened to. Alternatively, it is also possible to dynamically determine these scrambled frequencies, by limiting the interference analysis to the only candidate frequencies, that is to say to the authorized frequencies actually accessible by the radar at the current instant, it is to say those that were selected by the previous selection step 54. The analysis is then performed by the radar in real time by proceeding, before the issuance of a score, to a targeted listening to a restricted set of these some frequencies. This last way of proceeding is known under the name Anglo-Saxon "Instantaneous Least Jammed Frequencies" (ILJF).
As a result, the fifth step 55 of the method according to the invention distinguishes three cases:

• If none of the candidate frequencies associated with the selected candidate scores are scrambled, none of the candidate scores are discarded.
• if all the candidate frequencies are scrambled, the retained score is that associated with the least scrambled frequency.
• if only certain candidate frequencies are scrambled, only candidate scores that are only associated with scrambled frequencies are discarded.

It should be noted that for real-time operating constraints, step 55 and following can be operably inserted into the GRR, since the ILJF instant listening method is very time-constrained.
According to the invention the sixth step 56 of the method is applied to the candidate scores which were not rejected after the fifth step 55. It constitutes the last selection step and consists in retaining only the candidate score associated with the least used candidate frequency in previous iterations. This score and the corresponding frequency form the pair (pointing, frequency) finally selected.
The test performed during this sixth stage, illustrated by the figure 11 , is based on the study of the histogram of the authorized frequencies emitted during the last laps of the antenna (for each authorized frequency percentage of scores transmitted at this frequency). In the illustration of the figure 11 we consider the case where only two candidate scores each associated with a single frequency are still retained (represented by segments of lines in solid lines).
According to the invention, if a frequency 111, corresponding to the candidate scores retained at the end of the preceding step, has a lower utilization rate than the other frequency selected (FIG. 11-a, in solid lines the frequencies candidates), this frequency is chosen. As a result the corresponding score (s) are retained and the other candidate scores are discarded. On the other hand, if all the frequencies corresponding to the scores selected have an identical utilization rate as shown in FIG. figure 11-b where two frequencies 112 and 113 are to have an equal utilization rate, then the selected frequency (113) is the one which is closest to the nominal frequency f ₀ (114 for the frequency 112, 115 for the frequency 113) of the candidate score with which it is associated. As a result, the candidate score associated with this frequency is retained and the other scores are discarded.
Finally, if several scores are associated with the selected frequency, the one selected will be the one for which the antenna direction is closest to the exit of its managed visibility domain, or which is equivalent, the score of which time is the closest.

At the end of the sixth step 56 of the method according to the invention, only one candidate score is finally selected. The characteristic waveforms of this pointing are transmitted to the radar pulse burst management (GRR) which will then produce the temporal sequencing of the transmission phase and the radar reception phase corresponding to this score. The candidate score is then either removed from the candidate clocking table (case of idle clocking), or kept in memory in this table but in a deactivated form (case tracking pointers, which are reactivated according to their period of time). program).

dans laquelle:

• Nb_{freq_aleatoire} est le nombre de fréquences de la plage du tirage aléatoire;
• E() est la fonction qui renvoie la partie entière d'un nombre;
• Tirage_uniforme(.) est la fonction qui renvoie un nombre entre 0 et 1 suivant une loi uniforme;
• Coeff_tirage est un coefficient réel entre 0 et 1, utilisé en exposant de la fonction Tirage_uniforme(.). On pourra prendre, par exemple, une valeur égale à 0.5 et obtenir ainsi un tirage selon une loi qui suit une décroissance uniforme (loi linéaire) comme dans le cas illustré par la figure 7. Alternativement, on pourra prendre, par exemple, une valeur égale à 0.25 et obtenir ainsi un tirage selon une loi qui suit une décroissance quadratique.
• i_freq est l'indice de la fréquence sélectionnée, les fréquences de la plage du tirage aléatoire étant indicées de 0 pour la fréquence la plus basse à (Nb_{freq_aleatoire} - 1) pour la fréquence la plus haute. La fréquence correspondant à l'indice i_freq correspondra à la fréquence f_min de la fenêtre attribuée au pointage.
  Comme le montre des simulations effectuées par la déposante, ce tirage aléatoire particulier contrairement à un tirage aléatoire équiprobable contribue à obtenir avantageusement une distribution homogène au cours du temps des fréquences autorisées. Il contribue ainsi à ce que le radar soit moins sensible à certains types de brouillages.

The sixth step of the method according to the invention is followed by a seventh step 57 which aims to feed the table of candidate scores from new pointing requests or to reactivate tracking pointing requests that will soon be again visible by the antenna.
This step realizes in particular the choice of the range of frequencies that are associated with each score, frequencies which, as has been said above, are chosen by implementing a particular random draw. This random draw of frequencies, is primarily to ensure the diversity of frequencies that will ultimately played by the radar, all points combined. It consists in determining the frequencies f _min and f _max of beginning and end of range.
The starting range frequency f _min is randomly selected from frequencies in the range sufficiently far from the security window to ensure the achievement of the frequency diversity constraint. However, the determination of the frequency f _min , which characterizes the beginning of the frequency range allocated to the score, does not result from a simple equiprobable random draw for which the probability of choosing, for f _min , a given frequency of the plane of frequency mentioned above is the same for all the frequencies of the draw zone of the first frequency. It actually results from a draw for which the probability of choosing, for f _min , a given frequency is a decreasing function of the relative position of this frequency in the draw zone of the first frequency, the lowest frequencies having more lucky to be chosen as the highest.
According to the invention, the decay law is established in such a way that the minimum frequency (f _min ) is drawn by setting aside the highest frequencies of the frequency domain, frequency lower than the maximum frequency F _max . Thus, a minimum limit frequency f is _set lower than the frequency F _max so that the draw area of the frequency f _min is in the range [F _min , f _{limit_drawn} ]. The random draw area [F _min , f _draw-limit ] is such that the probability of drawing a frequency beyond f draw- _limit is zero. The law of the probabilities of drawing in the interval [F _min , f _{limit_drawing} ] of the minimum frequency f _min is also decreasing, which has advantageously the effect of compensating for the fact that the intersection of all the sets, which will ultimately be attributed to the scores, will favor the occurrence of selection of high frequencies in these sets.
The decreasing character of the f _min pulling law which favors the low frequencies advantageously compensates for the induced effect of systematic favoring of the high frequencies naturally caused by the frequency selection mode used during the steps of the method according to the invention. invention. It thus makes it possible to establish an equiprobability of the frequencies played by the radar.
According to the invention, the formula for drawing the minimum frequency f _min is given by the expression: $${\mathrm{i}}_{\mathrm{freq}}\mathrm{=}{\mathrm{Nb}}_{\mathrm{freq\_aleatoire}}\mathrm{-}\mathrm{1}\mathrm{-}\mathrm{E}\left({\mathrm{Nb}}_{\mathrm{freq\_aleatoire}}\mathrm{\cdot }\mathrm{Tirage\_uniforme}{\left(\mathrm{.}\right)}^{\mathrm{Coeff\_tirage}}\right)$$

in which:

• _{Random_freq} is the number of frequencies in the random draw range;
• E () is the function that returns the integer part of a number;
• Uniform_draw (.) Is the function that returns a number between 0 and 1 according to a uniform law;
• Coeff_tirage is a real coefficient between 0 and 1, used by exponent of the function Tirage_uniforme (.). We can take, for example, a value equal to 0.5 and thus obtain a draw according to a law which follows a uniform decay (linear law) as in the case illustrated by the figure 7 . Alternatively, we can take, for example, a value equal to 0.25 and thus obtain a draw according to a law that follows a quadratic decay.
• i _freq is the index of the selected frequency, the frequencies of the random draw range being indexed from 0 for the lowest frequency to (Nb _{freq_aleatoire} - 1) for the highest frequency. The frequency corresponding to the index i _{freq will} correspond to the frequency f _min of the window assigned to the score.
  As shown by the simulations performed by the applicant, this particular random draw, unlike a random draw equiprobable contributes advantageously to obtain a homogeneous distribution over time of the authorized frequencies. It thus contributes to the radar being less sensitive to certain types of interference.

Thus, step 57 of the method according to the invention makes it possible to associate with each of the candidate scores, a frequency range whose starting frequency, f _min , is determined so as to induce a more homogeneous distribution of the frequencies that will be set. implemented. These frequencies constitute the zone named "zone of draw of the first frequency". A given number of frequencies is thus assigned to each candidate score at the time of its integration into the table. The determination of the frequency range associated with each candidate score is advantageously carried out completely independently of one score to another. Each frequency determines, taking into account the position of the antenna and its rotational speed, a possible time of realization of the pointing considered. As illustrated by figure 7 , the frequency range thus determined thus makes it possible to define for each pointing request an angular zone of variable size (random) called the managed visibility area which when traversed by the antenna direction allows the radar to implement the matching score using one of the frequencies in the range. In this way, as illustrated by figure 8 , it is possible to determine, for a given direction of the antenna, the candidate scores 81 and 82 likely to be implemented at a given instant, since the antenna azimuth is in at least one of the visibility of candidate scores.

Subsequently, the requests for scores that have been chosen and those whose implementation is no longer possible (antenna direction located after the visibility domains of these requests) are eliminated from the table while the others remain there ( antenna direction located before or in the areas of visibility of these requests) so as to be taken into account at the next iteration.

In addition to choosing the range of frequencies that are associated with each candidate score, step 57 also has the function of constituting dynamically at each iteration, as illustrated by FIG. figure 12 , new pointing requests to take into account. This news queries are both new standby watch requests whose directions are included in a first angular window 121 contiguous to the field of view 122 of the radar (full visibility range if all frequencies are allowed or partial visibility area if only one part of the frequencies is allowed) as well as requests for other types of punishments that may occur, for example pursuit requests, queries whose directions are included in a second angular window 123 contiguous to the first window 121. The position of the Antenna 31 taken for origin is that obtained after taking into account the last score chosen.
According to a preferred embodiment of the invention, the size of the first angular window 121 is determined so as to correspond to the angle of rotation of the antenna over a period equivalent to the maximum duration of one half. a score to which is added the maximum delay that may exist between the moment when a score was selected and the moment when it is actually issued.
According to this preferred embodiment, the size of the second angular window 123 is in turn determined as being proportional to the azimuth extension value DAz _{AdaptationCharge} which corresponds to a multiple of the azimuthal extension of the partial visibility domain in the limit of the full visibility range. This window may be for example equal to 0.5 times the azimuth extension value DAz _{AdaptationCharge} . The theoretical limit beyond which a local overhead focused on this sector can no longer be treated, is in turn equal to twice the partial visibility area.
In practice, to dynamically create a standby watch in a given direction, included in the first window 121, the method according to the invention takes into account the pointing requests concerning scores other than standby scores whose direction is included in the second window 123 and the last standby pointer created at the previous iteration. As a result, the standby point of view considered is directed to the direction of the previous standby pointing, to which is added an azimuth deviation Difference _Standby defined and calculated as follows.

The implementation of this logic of adaptation of the radar to the load is made considering that the azimuthal deviation imposed by the process according to the invention between two neighboring watchpoints, EcartAz _Veille , comprises a long-term component, EcartAz _VeilieLT , and a short-term component, EcartAz _VeilleCT .
EcartAz _VeilleLT is very little affected by local load differences. Its value corresponds to the sector in azimuth swept by the antenna during the average duration of a score, duration calculated over a period of time of several seconds, taking into account a percentage of technical scores fixed a priori.
EcartAz _VeilleCT responds only to local load variations. Its value is given by the following expression: $${\mathrm{EcartAz}}_{\mathrm{VeilleCT}}={\mathrm{DAz}}_{\mathrm{Charge}}\cdot {\mathrm{EcartAz}}_{\mathrm{VeilleLT}}/\left({\mathrm{DAz}}_{\mathrm{AdaptationCharge}}-{\mathrm{DAz}}_{\mathrm{Charge}}\right)$$

In which DAz _Charge corresponds to the observed instantaneous radar load difference, reduced to a positive azimuth value in case of overload and negative in case of underload.
It can be seen that EcartAz _VeiueCT is positive in the event of an overload (the standby times are no longer set aside for a certain period of time so as to gradually release the radar load), negative in the case of _underloading (the standby times are tightened during a certain period of time). time to gradually take advantage of the surplus radar load available) and zero when the load is normal. In case of large and sudden local overload, the maximum spacing between punches by the process will be limited.
So that the difference in azimuth remains almost constant in the treated area, the calculation of EcartAz _VeineCT is carried out only in case of detection of a new cause of overload or under load (as for example the taking counting a tracking point or a variation of the antenna rotation speed) or when the state of charge has returned to normal. In the latter case, the difference in azimuth is considered to be given by the component EcartAz _VeilieLT .

Calculation of the azimuthal deviation thus calculated makes it possible to ensure that a state of radar overload or underload is best treated in an azimuthal sector of width equal to DAz _{AdaptationCharge} . In this sector Azimuth deviations between watch-points will be almost constant and the entry and exit of this sector will be without abrupt variation of these deviations, guaranteed in this by the sufficient width of the sector in question.
For each standby score to come, stored in the table of the candidate scores, the corresponding pointed direction is thus determined.

• si le pointage candidat est de veille il est purement et simplement supprimé de la table des pointages candidats,
• si le pointage candidat n'est pas un pointage de veille mais qu'il est cependant périodique (pointage de poursuite par exemple) il est conservé dans la table des pointages candidats mais l'état désactivé (il n'est pas pris en compte dans les étapes de sélection) et il sera de nouveau activé et ses paramètres seront alors réactualisés (comme par exemple sa position et son domaine de visibilité géré) au tour d'antenne suivant ou à sa période suivante lors de son nouveau passage dans la fenêtre angulaire 123.

The seventh step 57 is finally followed by a final step 58 of processing candidate scores that have not been selected in the last iterations and are no longer possible candidates for the next iteration because the direction of the antenna has left their respective managed visibility domains, two cases must be taken into account:

• if the candidate score is standby it is purely and simply deleted from the table of candidate scores,
• if the candidate score is not a standby score but it is however periodic (tracking pointing for example) it is kept in the table of candidate scores but the disabled state (it is not taken into account in the selection steps) and it will be activated again and its parameters will then be updated (for example its position and its field of visibility managed) at the next antenna revolution or at its next period when it is again in the angular window 123.

It should be noted that if, after an iteration of the various steps of the method according to the invention, no pointing is selected (detection of "dead time"), a technical score is implemented (cf. figure 13 ). A new iteration is then initiated.
In the same way, if the candidate scores are insufficient in the first step of a given iteration, a technical score is selected. The treatment of an insufficient number of candidate scores would not allow an efficient management of the frequencies, ie a maintenance of the equiprobability of the frequencies. A new iteration is then initiated.

The principle of managing "dead times" by inserting technical scores is illustrated in figure 13 .
If at any given time t ₁ , no active candidate score is available, either whereas, given the azimuth of the antenna, the visibility ranges, 134, 135 or 136, of available candidate scores 131, 132 or 133, are not accessible by the antenna beam at time t ₁ , that part of the field of view 138 of one or more available candidate scores 137 corresponding to the frequencies associated with these scores is not accessible by the antenna beam at time t (see FIG. figure 10 which shows examples of pointing accessible by the antenna due to the tolerance materialized by the window 103), a technical score 139 of a given duration is inserted. Then we determine the end time t ₂ of completion of the technical score and we examine whether at this moment, taking into account the rotation of the antenna, some of the preceding candidate scores, 131, 132, 133 or 137, have become assets.
If yes, the active candidate scores are taken into consideration.
If no, no candidate score is yet available, another technical score 1311 is inserted.
The operation is repeated until for a given time t ₃ , given the azimuth of the antenna at this time, a first candidate score 132 becomes in view of the direction of the potentially accessible antenna.
It should be noted that, according to the invention, the technical score rate is regulated via the function 57 (the difference between scores takes into account a percentage of time for the technical scores defined a priori and fixed by the operator) .

By using the dispersive nature of the antenna used by the radar in which it is implemented, the method according to the invention advantageously makes it possible, with regard to the pointing to be performed during the period of rotation of the antenna, to substitute for the notion of execution time the notion of interval of execution time. It thus makes it possible to optimally manage the radar load by determining the order of execution of the scores to be executed in a given time interval (which it has itself created), and makes it possible to insert addressed scores (such as tracking points) to turn this radar into a multifunction radar.

It makes it possible to harmoniously manage the variations of the radar load (caused by the insertion of tracking points and by the fluctuation of the antenna rotation speed) by adapting the spacing of the watch scores to these variations of charge.
It also makes it possible to effectively manage the frequencies, by playing randomly and homogeneously the authorized or least scrambled frequencies. With respect to interference, the method makes it possible to adapt to the scrambling card maintained in turn and instantaneous listening of the least scrambled frequencies, to select the best frequencies to use.

The possibility of multifunction type operation is available thanks to the ability to dynamically insert addressed pointers of the "active tracking", "Non-Cooperative Target Recognition (RNCC)", "external designation", ...; but also advanced anti-jam functions "Pick-A-Boo" (spacing scores around a scrambled direction, to limit the effects of jamming). For an RNCC pointing capability with broadband waveforms (ramp frequency modulation), and avoid misalignment of the beam with respect to the position of the target, the pointing will be scheduled by the method of the invention in several under reduced frequency ramps (principle of multi-ramp waveforms).

Claims (9)

1. A method for managing transmitted beam directions by a radar comprising a rotating dispersive antenna, the rotation speed of which is likely to vary over time, said management being carried out as a function of the field of visibility of the antenna at the considered instant, said method being applied in an iterative manner to candidate beam transmission directions, each candidate beam transmission direction being associated in a table to a set of candidate frequencies that are determined from the set of authorised radar frequencies, said method consisting in selecting, at each iteration, which of the candidate beam transmission directions has to be executed first, as well as the frequency that is associated with this beam transmission direction for its execution, characterised in that it comprises the following steps:

   - a step (51) which involves eliminating for the considered iteration the candidate beam transmissions which are not accessible to the antenna lobe, taking into account the frequencies associated therewith in the table and the beam transmission direction in which the antenna is pointed;

   - a step (53) which involves eliminating for each of the remaining beam transmission directions the associated candidate frequencies that are not located in a given neighbourhood of the theoretical frequency allowing the considered beam transmission to be executed in the required beam transmission direction, taking into account the beam transmission direction in which the antenna is pointed;

   - a step (55) which involves determining whether some of the remaining frequencies are jammed and in the case that all of the remaining frequencies are jammed the least jammed frequency is determined, the jammed frequencies being eliminated with the exception of the least jammed frequency;

   - a step (56) which involves selecting the least used frequency from the remaining candidate frequencies, with the remaining candidate beam transmission directions that are not associated with this frequency being eliminated;
   wherein the non-eliminated candidate beam transmission direction to which the least used candidate frequency is associated is transmitted to means that are designed to implement the corresponding waveforms;

   - a step (57) of creating new beam transmission direction requests, each new beam transmission direction request being constituted by a new candidate beam transmission direction or a candidate beam transmission direction that is not yet executed and a set of associated candidate frequencies, said frequencies being those of the authorised frequencies that are contained in a frequency band that is determined in a random manner from the frequency band B assigned to the radar;

   - a final step (58) of processing the case of beam transmission directions that are no longer visible on completion of the current iteration, taking into account the displacement of the field of visibility following the rotation of the antenna;

   the steps of the forming method being repeated in an iterative manner, with the table that associates the candidate beam transmission directions and the candidate frequencies being renewed at each cycle.
2. The method according to claim 1, characterised in that it further comprises an intermediate step (52) that consists in eliminating the candidate beam transmission directions that, if they were ultimately selected during the considered iteration, could induce the loss of one or more other candidate beam transmission directions with a short duration of visibility.
3. The method according to claim 2, characterised in that it further comprises a second intermediate step (54) for carrying out the selections of the beam transmission directions that are declared as having the highest priority.
4. The method according to any one of the preceding claims, characterised in that the step (57) of creating new beam transmission direction requests undertakes for each candidate beam transmission direction the association of a frequency range that is within an interval that is limited by two frequencies f_min and f_max, the frequency f_max being the highest frequency of the authorised frequency field, the frequency f_min being a frequency that is selected randomly in a frequency field that extends from the lowest frequency of the authorised frequency field to a frequency f_{limite_tirage} that is greater than f_min and lower than f_max.
5. The method according to claim 4, characterised in that the frequency f_min is obtained by a random draw with a likelihood density that decreases as the frequency increases.
6. The method according to claim 4 or 5, characterised in that the frequency f_{limite_tirage} is determined so as to define with f_max a frequency interval that contains a number n of authorised frequencies that is low in relation to the number of authorised frequencies contained in the interval that is limited by f_min and f_max.
7. The method according to any one of the preceding claims, characterised in that, for determining new candidate beam transmission directions, step (57) for creating new beam transmission direction requests takes into account a first angular window (121) that is contiguous with the field of visibility (122) for determining new monitoring beam transmission directions and a second angular window (123) that is contiguous with the first window (121) for determining other new beam transmission directions, the angular windows (121) and (123) being determined so as to take into account the rotation of the antenna.
8. The method according to claim 7, characterised in that the first angular window (121) is determined so as to correspond to the angle of rotation of the antenna over a duration that is equivalent to the maximum duration of a half-beam transmission direction, to which is added the maximum delay that can exist between the moment at which a beam transmission direction has been selected and the moment at which a beam is effectively transmitted, the size of the second angular window (123) being, for its part, defined as being proportional to the value of azimuth extension DAz_{AdaptationCharge} that corresponds to a multiple of the azimuth extension that corresponds to the partial field of visibility, DAz_{AdaptationCharge} being in all cases lower than the complete field of visibility.
9. The method according to any one of the preceding claims, characterised in that the step (55) takes into account the information that relates to the least jammed frequencies for selecting beam transmission directions in two possible manners, either locally by instantaneously listening to the jammed frequencies or globally by using the available jammed frequency maps.

Images